The new top quality badge

Discussion started by LemonadeCG

So what is this new badge? Who and on what criteria decides which model is given the badge and which is not? What sellers can do to earn it for their models? Do buyers get additional guarantees buying models with the badge? Cgtrader, please shed some light on the new feature, because at the moment it looks to me completely arbitrary, given at random and probably by some automated system. I would like to know what's the point in it? Thank you!

Answers

Posted 9 months ago
0

I don't see any new badges. Can you please tell how it looks? Is it like "pbr", "lowpoly" etc?

LemonadeCG wrote
LemonadeCG
Just go here and look for the small yellow badge on the top right corner of some of the models: https://www.cgtrader.com/3d-models
Posted 9 months ago
0

Oh, sorry, it appeared after clearing the cache.

Posted 9 months ago
0

I do not see the badges. I tried to clear the cache and watch from another browser. nothing but discount badges.
I feel myself ..... special.

Posted 9 months ago
1

I suspect CGT is still testing this new feature, because the badges appear and disappear regularly. I'm also curious how it will work.

Posted 9 months ago
1

Perhaps if model price is 3-5$ and "designer" uploaded it in packet of 300 another models using "batch upload" it decides "TOP QUALITY" badge.)))))

...shortly: another "genial" and "brilliant" fart from marketing "specialists".

Posted 9 months ago
3

@skapricorn, let's not be so negative and jump into conclusions without hearing a word from cgtrader first. I had discussion about this feature with a fellow designer and we both agree, that models inspected for quality and approved by the marketplace, could be a big selling point. The problem is, that model inspection/moderation takes a lot of time and human resources. Let's say it takes 5 minutes per model, which is not that much to properly assess its quality, then to inspect whole CGT database would take 5 millions minutes, or 10 years of work. Not very realistic, isn't? What other options then? Either to pick only very small portion of the models, or make quality assessment process automated. Somehow i don't see how either of these options could work, because former would be very unfair to the sellers, while latter would be like cheating to the customers. But maybe cgtrader has something different in mind and it would be great to hear their reasonings.

skapricorn wrote
skapricorn
Just want to add: CGtrader obviously has MANY technical problems and issues (may be it's servers overload or bad quality of code etc). Instead of solving tech.side CGT decided to "develop" marketing. :)
Posted 9 months ago
0

Maybe there will be more rather automated factors.
I tried to expect some models categories - for example "trains" under "best match" sorting - and first 4 models with this new batch have 8, 4, 4 and 2 reviews - so probably amount of positive customers reviews has big impact on it.

zabotlama wrote
zabotlama
correction: inspect not expect :-D
LemonadeCG wrote
LemonadeCG
I'm not sure about that. Some of my big sellers with lots of positive reviews, didn't get the badge, while some fresh models with zero sales and reviews, got it. I tried hard to see some pattern here, but i couldn't find one.
Posted 9 months ago
0

So, who knows. Maybe after some testing period we will get some official info.

Posted 9 months ago
0

I must conclude, it's done in typical cgtrader fashion - let's introduce a major feature on friday and leave the marketplace unattended over weekend, find out how chaos will evolve :]

Posted 9 months ago
1

Not wanting to add to rumours without actually knowing something … but it's definitely an automated process. I uploaded a new item and it immediately had the badge. Most of my previous items have the badge, but the ones lower on my list, ie the oldest loads, do not. Presumably the automated process hasn't got that far yet.

I know when we see the words "Top Quality" we assume someone sitting there judging the worth of the model, but that would be a ridiculous waste of resources. And to what criteria? And who judges what is quality anyway? I suspect it's something far, far simpler, maybe just a check that the item loads without errors.

Posted 9 months ago
2

Hi all,

Glad to see discussion going on. We're receiving a lot of suggestions asking to highlight higher quality 3D models that could be offered to buyers with some kind of quality proof or distinction (aka badge).

As some of you correctly identified, manual inspection would take years and implementing automated process would also require a lot of work before system is fully trustworthy and functional.

For this reason, we're running a low scale test to see if highlighting higher quality items with a badge could bring more attention and result in more sales, if implementation is right. This is a temporary test that would help us to decide if we should invest more into developing the feature further. In this case, we used a collection of undisclosed factors that indicate higher quality of a 3D model.

I'd like to also add that we're analysing and solving marketplace performance issues to accommodate it's growth. Your insights and suggestions are always welcome and taken into account.

Feel free to ask questions and thanks for sharing your concerns.

Best,
Gytis from Marketing Team

Posted 9 months ago
5

I understand the general thinking behind this, I'm not sure about it's wisdom though. None of this is criticism, so please take it as constructive discussion.

Firstly, I don't see it generating new sales or customers. At best it might shift the customers focus away from an unbadged item and onto a badged item, but the number of sales will still be the same, the customer will just be spending their money on different items. So the net gain here seems to be zero.

The wording on the badge is "Top quality approved by CGTrader". This implies very strongly that an inspection process has been conducted on the actual item, and it has met certain quality criteria. These criteria must be very well thought out if the customer is to trust it. The criteria must only be about the model itself, not anything to do with sales, views, likes, etc, or even the reputation of the designer. The criteria must be entirely from some sort of examination (manual or automated) of the actual item. Even the best designers sometimes make mistakes, like forget to load a texture, or something. Personally I'm struggling to understand what sort of automated process can judge quality with this level of assurance, but I'm happy to defer that to cleverer people, and accept that it is theoretically possible to have such an automated process … I'm just sceptical for the moment!

At the moment approximately half my items have the badge and half do not. I'm hoping this is because this is a "limited run" test and not the fully blown process. All of my items are made using the same technique and tools and they are all of a very similar likeness because of the niche market I'm catering to. In other words I would expect either ALL of my items to be badged or NONE of them to be badged. As it currently sits with half/half, it strongly implying the process is not working or is flawed in it's criteria.

I also have something of an issue with what will happen to items that don't receive the badge. A perfectly good, well made item may be loaded, but for whatever reason it doesn't capture the customers imagination and ends up being a non-seller. No fault of the model or the designer, just the whims of the marketplace. If the item is also unbadged, then it will likely remain a non-seller forever as customers will incorrectly judge it to be flawed.

Then there is the reputation of CGT to take into account. If the automated process fails for some reason and assigns a badge to an item with an issue, then later a customer downloads the item and finds it flawed. This would immediately cause them to question the value of the badge in the first place, and unlikely for them to trust it again in the future. And CGT would have a damaged reputation because they endorsed something that was flawed. In other words, the process that assigns the badges has to be absolutely bomb-proof.

Okay, sorry for the long post, hope it was of some value.

everyplant wrote
everyplant
Also, another observation I would like to add to my original post, which will need to be addressed before a full blown system is rolled out. Some of my collections have the badge, but the individual models that make up the collection do not. So what does this mean? That the collection is top quality, but the models are not? And also the opposite, all the individual models making up a collection have the badge, but the collection itself does not. A very strange inconsistency if the models have been examined for "quality".
Posted 9 months ago
4

I think the same as @everyplant and also noticed inconsistency in the selection of the models.
I don't want to show examples here, but I saw models with a badge that are not really bad, but I wouldn't consider as "top quality" and also really good models that don't get the badge.
Semi-randomly selecting only part of all models, just for the sake of marketing can have a negative effect.

Posted 9 months ago
2

Looks as inconsistent as the other "approvals" here. Many animated and even rigged models are on here which don't even get the "badges" "rigged" or at least "animated" although they are presented with animation / video. Some get the badge "rigged" even though there is no rig seen in the presentation.
It's interesting how automated processes can figure that all out / what are the criteria they work with?

Posted 9 months ago
4

I completely agree with what has already been said. The badge allocation process must be thorough and reliable because CGT risks your credibility and reputation. Putting the quality seal without verifying that it is true, is like issuing a certificate without knowing its content, or signing a contract without having read it (stupid). Automated allocation can be circumvented, as is already the case with model scoring (models with 1 presentation image repeated 100 times, model description with entire articles from Wikipedia ...). This causes it to completely lose its value. It is useless and will also negatively affect We already have several of those useless badges (rigged, animated)

Even though automated tools are used for some checks, human intervention is required to check that everything is correct. This does not have to take up too much time per model, for an expert eye, (good modelers have a clear, consistent style, with clean meshes ...). Nor should it be applied to the million models (only in free models, more than 225,000 would be eliminated)

The evaluation criteria should refer exclusively to the model: quality, presentation and price (views, votes, comments ... are also very manipulated by the clone armies that many users have). Mainly the quality of the model should be evaluated, but if you want the highest quality you should also have a good presentation of the model: detailed description, complete model details, clear images of the mesh, UV map, textures, armature, poses ..., (the most published models lack these details.) They must have a native format, with software updated to the date of publication (many of the models that are published today, do so with software from 10 years ago, or without native format). And also another very important point is that they must have a fair price (quality is not cheap, because it involves many hours of dedication). There are many models that do not meet any of these criteria, so by this point, the number of models to examine would also drop considerably.

The quality of the model unquestionably begins with the quality of the mesh, (100% quads manifold meshes, without intersecting or distorted faces, without poles of more than 5 edges, with loops appropriate to the shape of the model, UV map without overlapping and not distorted, original textures of standard size ...) All these details cannot be falsified and are easily quantifiable (any program includes tools to verify this data, and it can be automated).

And above all, I would avoid certifying: stolen models, models made with pirated software, counterfeits, printable models that are not, those that violate copyrights, trademarks, patents, or any other law, because doing so is like affirming that you know about these violations and that you agree (something that can exploit in your face), because a certification is a guarantee, in this case of quality and is also binding. Many buyers may be deceived by not receiving the expected quality and this could lead to returns and the loss of quality seeking customers (which are the most rare). At this point, a large number of models, not suitable for certification, would be eliminated.

It is a good opportunity, so that those who sell their models for very low of their real value, are encouraged to upload it. That those who make bad presentations and do not show the mesh, textures ... are encouraged to do so and to reward those who are original and really do things well.

Do it right or don't do it (for now you haven't done very well). If you do it right you will attract quality buyers, willing to pay more, for authentic, quality models, but if you do it wrong, you will lose credibility and time (time that you could use to solve real problems already known)

Mantas-Talmantas wrote
Mantas-Talmantas
There are cases where ower laping uv's are much prefered (some game engines to conserve texture size in gpu) i also think that triangular mesh is better in some cases (like sphere made out of equal size triangles better then quads being big on equator ant tine in poles for some physics simulation engines... ) So your set of rules is biased favoring animation / vfx industry...
everyplant wrote
everyplant
Herein lies the danger of defining what actually is "quality", and who sets the standard. I agree that all of your points are valid and represent good design practice, but I disagree that they can be used as a set of "musts" to make an item quality. The absence of a native format is no indication of the quality of the item, the model may well still be well made and of very high quality. For example, I don't include native formats with my models because I make them in a procedural editor that is so obscure it wouldn't be useful to anyone, and the format doesn't even appear in the dropdown. Also, the age of the software that is used to make it doesn't determine the quality of the final product, I've seen extraordinarily talented people make stuff in Wings3D that people with mega thousand dollar software couldn't compete with. Non-overlapping UV should be the aim where possible, but not always feasible … imagine how many thousands of texture maps you would need to have if modelling a tree with 1M+ leaves and you couldn't use overlapping textures for the leaves, no software on earth will load it without crashing! The argument for the use of tris has already been made. And so on. Please don't misunderstand me, you're points are all very good and I agree with them as being appropriate design principles and quality guidelines, but I don't believe you can set them as hard and fast rules to determine an items quality.
Posted 9 months ago
0

Game engines, libraries like webgl and others ... use triangulated meshes because they greatly simplify the calculations. A triangle by definition has all its points on the same plane, (the normals are parallel) and any geometric shape can be decomposed into triangles. But buyers are people, and people work best with quads and loops (by definition a loop ends when it reaches a triangle). Working with loops, facilitates the selection, creation or elimination of new loops, you can easily modify your flow ... in short, it will facilitate the life of the buyer to be able to easily adapt the mesh to their needs. And once you have it to your liking you can triangulate it if necessary, all programs have an option to do this. A 100% quads mesh without distorted faces will always have twice as many triangles as quads if it is triangulated and the process could be easily reversed. But in a mesh with faces with 4 vertices that are distorted, each quad can be divided into many more than 2 triangles and if the process is reversed, the mesh would be destroyed. Having overlapping on the UV map can also cause problems with some shaders and it is always easier to have it neatly arranged and stacked, than to have it stacked and have to be separated.

There can always be exceptions, such as in your case, or for scanned or printable models. But the truth is, most of the models that do not have a native format are because they have been downloaded from a pirated page and many of the triangulated meshes come from models extracted from games ... Having access to the native format will always allow the buyer get the most out of the model. The software frequently makes changes to the formats between versions. A professional has to be up to date in order to offer his clients the best, there are many things (rig, particles ...) that have evolved a lot in 10 years).

The criteria used must be objective and quantifiable, for example if it is used as a criterion that does not have triangles, or if it is fulfilled or not, there is no place for interpretations. This facilitates automation and is more fluid because each person can check for himself whether the model meets the criteria.

azunyan wrote
azunyan
The model for game engines with baked normals should be triangulated. And overlapping uvs allow to achieve higher texel density.
Mantas-Talmantas wrote
Mantas-Talmantas
also on native format. so i do 3d scanning and retopo for games. what is my native format? a 200 or thousand .jpg images from camera (or .dng) ? or is it realitycapture (photogrammetry software) project that can be ~ 20 gb in size and will opened by no one in game industry that my models are for? so i mark it as 3d max just cos i do some parts of my model preparation there but its not "true native" the result is .obj or .fbx and these are not called "native"
Posted 9 months ago
0

Game engines, libraries like webgl and others ... use triangulated meshes because they greatly simplify the calculations. A triangle by definition has all its points on the same plane, (the normals are parallel) and any geometric shape can be decomposed into triangles. But buyers are people, and people work best with quads and loops (by definition a loop ends when it reaches a triangle). Working with loops, facilitates the selection, creation or elimination of new loops, you can easily modify your flow ... in short, it will facilitate the life of the buyer to be able to easily adapt the mesh to their needs. And once you have it to your liking you can triangulate it if necessary, all programs have an option to do this. A 100% quads mesh without distorted faces will always have twice as many triangles as quads if it is triangulated and the process could be easily reversed. But in a mesh with faces with 4 vertices that are distorted, each quad can be divided into many more than 2 triangles and if the process is reversed, the mesh would be destroyed. Having overlapping on the UV map can also cause problems with some shaders and it is always easier to have it neatly arranged and stacked, than to have it stacked and have to be separated.

There can always be exceptions, such as in your case, or for scanned or printable models. But the truth is, most of the models that do not have a native format are because they have been downloaded from a pirated page and many of the triangulated meshes come from models extracted from games ... Having access to the native format will always allow the buyer get the most out of the model. The software frequently makes changes to the formats between versions. A professional has to be up to date in order to offer his clients the best, there are many things (rig, particles ...) that have evolved a lot in 10 years).

The criteria used must be objective and quantifiable, for example if it is used as a criterion that does not have triangles, or if it is fulfilled or not, there is no place for interpretations. This facilitates automation and is more fluid because each person can check for himself whether the model meets the criteria.

The quality criteria must be demanding, because if it is not, ALL or most of the models will have the badge, with which it is as if none had it.

Posted 9 months ago
0

A mesh of quads, can be easily modified (create new loops) and can also be easily triangulated in any program. Therefore, it can meet the needs of buyers who want a quad mesh and those who want a triangulated mesh. The number of potential buyers is greater. I don't see the need to sell it triangulated.

The criteria that I have proposed are for general cases. Scanned and printable models are special cases that will need special criteria. For example in TS, I don't know how those special cases will be treated, but the criteria they use for the checkmate are not very different from what I have proposed. And they have years of advantage in this matter, they have developed the procedure listening to the proposals of their clients, who are professionals in the industry and after all are the ones who have the last word.

azunyan wrote
azunyan
Different software can triangulate differently. So if normal map was baked on mesh with one triangulation variant applying same normal map on mesh with different triangulation can cause shading issues.
Posted 9 months ago
2

Thanks a lot for your input and constructive feedback.

I see quite many of you are concerned about the quality inspection mechanism and how it would work actually. And it's quite sensible/subjective aspect as many of you correctly indicated.

At this point, our goal is to understand the benefit/bottleneck scope of introducing a broader update. This is the exact reason why we're running the test (and many more) in the first place, as it tells us if we're moving to the right direction. In a case test indicators are positive, the question how to develop reliable quality inspection mechanism will be addressed. And this is there your feedback is really handy from designers (sellers) perspective.

Just to clarify, the test will close in short and is only applicable to small portion of 3D models.

gytiscg wrote
gytiscg
UPDATE: the test is actually closed.

Your answer

In order to post an answer, you need to sign in.

Help